The proposed research will seek to utilize population genetic data to distinguish regions of the human genome experiencing purifying selection from unconstrained genomic regions. Because genomic sequences subject to selective constraint perform functions beneficial to the organism, this work will reveal previously unknown functional regions of the human genome. In particular, since this approach does not rely on comparisons between humans and closely related species, it can uncover regions acquiring or losing selective constraint after humans split from other great apes. Regions acquiring function during this time period would represent an important class of recent human adaptations, and could reveal molecular changes responsible for uniquely human phenotypes. Beyond its evolutionary importance, this work would improve the functional annotation of the human genome, revealing functional regions that could result in harmful effects if disrupted, and that cannot be detected from comparative genomic techniques. In addition to revealing human-specific gains-of- function, the proposed project would allow for detection of losses-of-function occurring since the human- chimpanzee divergence. These events could also underlie important phenotypic changes in recent human evolution, as several known human-specific losses-of-function were adaptive. Even fitness-neutral losses of function are informative, as they may reveal differences in selective pressures allowing certain functions to be lost in humans but requiring them to be maintained in our relatives. Finally, the work proposed here will combine population genetic and phylogenetic data to reveal constrained regions with better accuracy than can be achieved by examining either of these types of data alone. This will result in further improvements to the functional annotation of the human genome, especially with respect to non-protein-coding functional regions that cannot be reliably detected by ab initio techniques. Performing this research will improve the applicant's knowledge of population genetics and computational methods that can leverage polymorphism to draw inferences about the selective and functional importance of different genomic loci. Instruction from a sponsor and co-sponsor with expertise in both of these areas, as well as interaction with other faculty members and postdocs at the sponsor's institution, will be invaluable for improving the applicant's skills. This experience wil greatly enhance the applicant's chances of achieving his goal of succeeding as an independent scientist running a lab at a research university.